Tester
Lipp Laboratory, Department of Environmental Health Science, University of Georgia
2020-11-24
Load all required libraries.
library(tidyverse)## Warning: package 'tidyverse' was built under R version 3.6.3## -- Attaching packages ---------------------------------------------------------------------------- tidyverse 1.3.0 --## v ggplot2 3.3.2 v purrr 0.3.4
## v tibble 3.0.3 v dplyr 1.0.0
## v tidyr 1.1.0 v stringr 1.4.0
## v readr 1.3.1 v forcats 0.5.0## Warning: package 'ggplot2' was built under R version 3.6.3## Warning: package 'tibble' was built under R version 3.6.3## Warning: package 'readr' was built under R version 3.6.3## Warning: package 'dplyr' was built under R version 3.6.3## Warning: package 'forcats' was built under R version 3.6.3## -- Conflicts ------------------------------------------------------------------------------- tidyverse_conflicts() --
## x dplyr::filter() masks stats::filter()
## x dplyr::lag() masks stats::lag()library(plotly)## Warning: package 'plotly' was built under R version 3.6.3##
## Attaching package: 'plotly'## The following object is masked from 'package:ggplot2':
##
## last_plot## The following object is masked from 'package:stats':
##
## filter## The following object is masked from 'package:graphics':
##
## layoutlibrary(broom)## Warning: package 'broom' was built under R version 3.6.3Read in raw data from RDS.
raw_data <- readRDS("./n1_n2_cleaned_cases.rds")Make a few small modifications to names and data for visualizations.
final_data <- raw_data %>% mutate(log_copy_per_L = log10(mean_copy_num_L)) %>%
rename(Facility = wrf) %>%
mutate(Facility = recode(Facility,
"NO" = "WRF A",
"MI" = "WRF B",
"CC" = "WRF C"))Seperate the data by gene target to ease layering in the final plot
#make three data layers
only_positives <<- subset(final_data, (!is.na(final_data$Facility)))
only_n1 <- subset(only_positives, target == "N1")
only_n2 <- subset(only_positives, target == "N2")
only_background <<-final_data %>%
select(c(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke)) %>%
group_by(date) %>% summarise_if(is.numeric, mean)
#specify fun colors
background_color <- "#7570B3"
seven_day_ave_color <- "#E6AB02"
marker_colors <- c("N1" = '#1B9E77',"N2" ='#D95F02')#remove facilty C for now
#only_n1 <- only_n1[!(only_n1$Facility == "WRF C"),]
#only_n2 <- only_n2[!(only_n2$Facility == "WRF C"),]
only_n1 <- only_n1[!(only_n1$Facility == "WRF A" & only_n1$date == "2020-11-02"), ]
only_n2 <- only_n2[!(only_n2$Facility == "WRF A" & only_n2$date == "2020-11-02"), ]Build the main plot
#first layer is the background epidemic curve
p1 <- only_background %>%
plotly::plot_ly() %>%
plotly::add_trace(x = ~date, y = ~new_cases_clarke,
type = "bar",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Daily Cases: ', new_cases_clarke),
alpha = 0.5,
name = "Daily Reported Cases",
color = background_color,
colors = background_color,
showlegend = FALSE) %>%
layout(yaxis = list(title = "Daily Cases", showline=TRUE)) %>%
layout(legend = list(orientation = "h", x = 0.2, y = -0.3))
#renders the main plot layer two as seven day moving average
p1 <- p1 %>% plotly::add_trace(x = ~date, y = ~X7_day_ave_clarke,
type = "scatter",
mode = "lines",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Seven-Day Moving Average: ', X7_day_ave_clarke),
name = "Seven Day Moving Average Athens",
line = list(color = seven_day_ave_color),
showlegend = FALSE)
#renders the main plot layer three as positive target hits
p2 <- plotly::plot_ly() %>%
plotly::add_trace(x = ~date, y = ~mean_copy_num_L,
type = "scatter",
mode = "markers",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Facility: ', Facility,
'</br> Target: ', target,
'</br> Copies/L: ', round(mean_copy_num_L, digits = 2)),
data = only_n1,
symbol = ~Facility,
marker = list(color = '#1B9E77', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_trace(x = ~date, y = ~mean_copy_num_L,
type = "scatter",
mode = "markers",
hoverinfo = "text",
text = ~paste('</br> Date: ', date,
'</br> Facility: ', Facility,
'</br> Target: ', target,
'</br> Copies/L: ', round(mean_copy_num_L, digits = 2)),
data = only_n2,
symbol = ~Facility,
marker = list(color = '#D95F02', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
layout(yaxis = list(title = "SARS CoV-2 Copies/L",
showline = TRUE,
type = "log",
dtick = 1,
automargin = TRUE)) %>%
layout(legend = list(orientation = "h", x = 0.2, y = -0.3))
#adds the limit of detection dashed line
p2 <- p2 %>% plotly::add_segments(x = as.Date("2020-03-14"),
xend = ~max(date + 10),
y = 3571.429, yend = 3571.429,
opacity = 0.35,
line = list(color = "black", dash = "dash")) %>%
layout(annotations = list(x = as.Date("2020-03-28"), y = 3.8, xref = "x", yref = "y",
text = "Limit of Detection", showarrow = FALSE))
p1## Warning: `arrange_()` is deprecated as of dplyr 0.7.0.
## Please use `arrange()` instead.
## See vignette('programming') for more help
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.## Warning: Ignoring 1 observations p2## Warning: `group_by_()` is deprecated as of dplyr 0.7.0.
## Please use `group_by()` instead.
## See vignette('programming') for more help
## This warning is displayed once every 8 hours.
## Call `lifecycle::last_warnings()` to see where this warning was generated.p_combined <-
plotly::subplot(p2,p1, # plots to combine, top to bottom
nrows = 2,
heights = c(.6,.4), # relative heights of the two plots
shareX = TRUE, # plots will share an X axis
titleY = TRUE
) %>%
# create a vertical "spike line" to compare data across 2 plots
plotly::layout(
xaxis = list(
spikethickness = 1,
spikedash = "dot",
spikecolor = "black",
spikemode = "across+marker",
spikesnap = "cursor"
),
yaxis = list(spikethickness = 0)
)## Warning: Ignoring 1 observationsp_combinedSave the plot to pull into the index
save(p_combined, file = "./plotly_fig.rda")Save an htmlwidget for website embedding
htmlwidgets::saveWidget(p_combined, "plotly_fig.html")Build loess smoothing figures figures
#create smoothing data frames
#n1
smooth_n1 <- only_n1 %>% select(-c(Facility)) %>%
group_by(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke) %>%
summarize(sum_copy_num_L = sum(mean_total_copies)) %>%
ungroup() %>%
mutate(log_sum_copies_L = log10(sum_copy_num_L)) %>%
mutate(target = "N1")## `summarise()` regrouping output by 'date', 'cases_cum_clarke', 'new_cases_clarke', 'X7_day_ave_clarke' (override with `.groups` argument)#n2
smooth_n2 <- only_n2 %>% select(-c(Facility)) %>%
group_by(date, cases_cum_clarke, new_cases_clarke, X7_day_ave_clarke, cases_per_100000_clarke) %>%
summarize(sum_copy_num_L = sum(mean_total_copies)) %>%
ungroup() %>%
mutate(log_sum_copies_L = log10(sum_copy_num_L)) %>%
mutate(target = "N2")## `summarise()` regrouping output by 'date', 'cases_cum_clarke', 'new_cases_clarke', 'X7_day_ave_clarke' (override with `.groups` argument)#add trendlines
#extract data from geom_smooth
#n1 extract
# *********************************span 0.6***********************************
#*****************Must always update the n = TOTAL NUMBER OF DAYS*************************
extract_n1 <- ggplot(smooth_n1, aes(x = date, y = log_sum_copies_L)) +
stat_smooth(aes(outfit=fit_n1<<-..y..), method = "loess", color = '#1B9E77',
span = 0.6, n = 161)## Warning: Ignoring unknown aesthetics: outfit#n2 extract
extract_n2 <- ggplot(smooth_n2, aes(x = date, y = log_sum_copies_L)) +
stat_smooth(aes(outfit=fit_n2<<-..y..), method = "loess", color = '#1B9E77',
span = 0.6, n = 161)## Warning: Ignoring unknown aesthetics: outfit#look at the fits to align dates and total observations
#n1
extract_n1## `geom_smooth()` using formula 'y ~ x'
fit_n1## [1] 11.70737 11.77145 11.83537 11.89845 11.96000 12.01935 12.07582 12.12872
## [9] 12.17874 12.22704 12.27370 12.31880 12.36241 12.40459 12.44544 12.48501
## [17] 12.52338 12.56063 12.59683 12.63205 12.66636 12.69985 12.73152 12.76046
## [25] 12.78687 12.81095 12.83288 12.85287 12.87111 12.88780 12.90312 12.91728
## [33] 12.93046 12.94286 12.95469 12.96612 12.97736 12.98861 13.00005 13.01188
## [41] 13.02429 13.03749 13.05166 13.06257 13.06659 13.06488 13.05858 13.04883
## [49] 13.03677 13.02354 13.01030 12.99817 12.98831 12.98185 12.97995 12.98373
## [57] 12.99435 13.01395 13.04290 13.07968 13.12276 13.17062 13.22173 13.27457
## [65] 13.32761 13.37933 13.42820 13.47270 13.51130 13.54249 13.56472 13.58341
## [73] 13.60467 13.62810 13.65326 13.67974 13.70712 13.73497 13.76288 13.79042
## [81] 13.81717 13.84271 13.86662 13.88847 13.90785 13.92434 13.93751 13.94695
## [89] 13.95222 13.95292 13.94862 13.93889 13.91970 13.88856 13.84734 13.79795
## [97] 13.74226 13.68217 13.61957 13.55635 13.49439 13.43560 13.38184 13.33503
## [105] 13.29703 13.26975 13.24713 13.22222 13.19557 13.16774 13.13930 13.11079
## [113] 13.08278 13.05582 13.03047 13.00729 12.98683 12.96966 12.95632 12.94738
## [121] 12.94217 12.93953 12.93930 12.94133 12.94545 12.95151 12.95934 12.96879
## [129] 12.97970 12.99191 13.00526 13.01958 13.03473 13.05054 13.06685 13.08350
## [137] 13.10034 13.11720 13.13393 13.15036 13.16748 13.18625 13.20647 13.22792
## [145] 13.25039 13.27368 13.29758 13.32187 13.34636 13.37113 13.39647 13.42241
## [153] 13.44897 13.47621 13.50416 13.53284 13.56231 13.59260 13.62374 13.65577
## [161] 13.68873#n2
extract_n2## `geom_smooth()` using formula 'y ~ x'
fit_n2## [1] 11.51689 11.61755 11.71729 11.81532 11.91088 12.00320 12.09151 12.17504
## [9] 12.25454 12.33131 12.40545 12.47705 12.54618 12.61293 12.67739 12.73965
## [17] 12.79978 12.85788 12.91403 12.96832 13.02083 13.07165 13.11968 13.16392
## [25] 13.20459 13.24191 13.27608 13.30733 13.33587 13.36192 13.38570 13.40741
## [33] 13.42728 13.44553 13.46236 13.47800 13.49266 13.50656 13.51991 13.53292
## [41] 13.54583 13.55884 13.57216 13.58029 13.57859 13.56856 13.55173 13.52961
## [49] 13.50372 13.47557 13.44668 13.41857 13.39275 13.37074 13.35405 13.34421
## [57] 13.34272 13.34854 13.35908 13.37363 13.39143 13.41175 13.43386 13.45701
## [65] 13.48046 13.50348 13.52533 13.54526 13.56255 13.57645 13.58623 13.59544
## [73] 13.60785 13.62306 13.64066 13.66026 13.68145 13.70382 13.72697 13.75050
## [81] 13.77399 13.79706 13.81928 13.84026 13.85960 13.87689 13.89173 13.90370
## [89] 13.91241 13.91746 13.91844 13.91494 13.90653 13.89343 13.87623 13.85550
## [97] 13.83183 13.80580 13.77799 13.74899 13.71937 13.68971 13.66061 13.63263
## [105] 13.60636 13.58239 13.55885 13.53366 13.50708 13.47938 13.45080 13.42162
## [113] 13.39208 13.36244 13.33297 13.30392 13.27555 13.24811 13.22188 13.19709
## [121] 13.17201 13.14490 13.11609 13.08592 13.05470 13.02276 12.99042 12.95802
## [129] 12.92588 12.89432 12.86367 12.83426 12.80640 12.78044 12.75668 12.73547
## [137] 12.71712 12.70196 12.69031 12.68251 12.67697 12.67204 12.66793 12.66485
## [145] 12.66300 12.66260 12.66384 12.66694 12.67210 12.67928 12.68824 12.69896
## [153] 12.71139 12.72549 12.74121 12.75851 12.77736 12.79772 12.81953 12.84276
## [161] 12.86736#assign fits to a vector
n1_trend <- fit_n1
n2_trend <- fit_n2
#extract y min and max for each
limits_n1 <- ggplot_build(extract_n1)$data## `geom_smooth()` using formula 'y ~ x'limits_n1 <- as.data.frame(limits_n1)
n1_ymin <- limits_n1$ymin
n1_ymax <- limits_n1$ymax
limits_n2 <- ggplot_build(extract_n2)$data## `geom_smooth()` using formula 'y ~ x'limits_n2 <- as.data.frame(limits_n2)
n2_ymin <- limits_n2$ymin
n2_ymax <- limits_n2$ymax
#reassign dataframes (just to be safe)
work_n1 <- smooth_n1
work_n2 <- smooth_n2
#fill in missing dates to smooth fits
work_n1 <- work_n1 %>% complete(date = seq(min(date), max(date), by = "1 day"))
date_vec_n1 <- work_n1$date
work_n2 <- work_n2 %>% complete(date = seq(min(date), max(date), by = "1 day"))
date_vec_n2 <- work_n2$date
#create a new smooth dataframe to layer
smooth_frame_n1 <- data.frame(date_vec_n1, n1_trend, n1_ymin, n1_ymax)
smooth_frame_n2 <- data.frame(date_vec_n2, n2_trend, n2_ymin, n2_ymax)#make plotlys
#plot smooth frames
p3 <- plotly::plot_ly() %>%
plotly::add_lines(x = ~date_vec_n1, y = ~n1_trend,
data = smooth_frame_n1,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n1,
'</br> Median Log Copies: ', round(n1_trend, digits = 2),
'</br> Target: N1'),
line = list(color = '#1B9E77', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_lines(x = ~date_vec_n2, y = ~n2_trend,
data = smooth_frame_n2,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n2,
'</br> Median Log Copies: ', round(n2_trend, digits = 2),
'</br> Target: N2'),
line = list(color = '#D95F02', size = 8, opacity = 0.65),
showlegend = FALSE) %>%
plotly::add_ribbons(x ~date_vec_n1, ymin = ~n1_ymin, ymax = ~n1_ymax,
showlegend = FALSE,
opacity = 0.25,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n1, #leaving in case we want to change
'</br> Max Log Copies: ', round(n1_ymax, digits = 2),
'</br> Min Log Copies: ', round(n1_ymin, digits = 2),
'</br> Target: N1'),
name = "",
line = list(color = '#1B9E77')) %>%
plotly::add_ribbons(x ~date_vec_n2, ymin = ~n2_ymin, ymax = ~n2_ymax,
showlegend = FALSE,
opacity = 0.25,
hoverinfo = "text",
text = ~paste('</br> Date: ', date_vec_n2, #leaving in case we want to change
'</br> Max Log Copies: ', round(n2_ymax, digits = 2),
'</br> Min Log Copies: ', round(n2_ymin, digits = 2),
'</br> Target: N2'),
name = "",
line = list(color = '#D95F02')) %>%
layout(yaxis = list(title = "Total Log SARS CoV-2 Copies",
showline = TRUE,
automargin = TRUE)) %>%
layout(xaxis = list(title = "Date")) %>%
plotly::add_segments(x = as.Date("2020-06-24"),
xend = as.Date("2020-06-24"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "Bars Repoen",
hoverinfo = "text",
text = "</br> Bars Reopen",
"</br> 2020-06-24",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-07-09"),
xend = as.Date("2020-07-09"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "Mask Mandate",
hoverinfo = "text",
text = "</br> Mask Mandate",
"</br> 2020-07-09",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-08-20"),
xend = as.Date("2020-08-20"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "</br> Classes Begin",
"</br> 2020-08-20",
hoverinfo = "text",
text = "Classes Begin",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_segments(x = as.Date("2020-10-03"),
xend = as.Date("2020-10-03"),
y = ~min(n1_ymin), yend = ~max(n1_ymax),
opacity = 0.35,
name = "</br> First Home Football Game",
"</br> 2020-10-03",
hoverinfo = "text",
text = "First Home Football Game",
showlegend = FALSE,
line = list(color = "black", dash = "dash")) %>%
plotly::add_markers(x = ~date, y = ~log_sum_copies_L,
data = smooth_n1,
hoverinfo = "text",
showlegend = FALSE,
text = ~paste('</br> Date: ', date,
'</br> Actual Log Copies: ', round(log_sum_copies_L, digits = 2)),
marker = list(color = '#1B9E77', size = 6, opacity = 0.65)) %>%
plotly::add_markers(x = ~date, y = ~log_sum_copies_L,
data = smooth_n2,
hoverinfo = "text",
showlegend = FALSE,
text = ~paste('</br> Date: ', date,
'</br> Actual Log Copies: ', round(log_sum_copies_L, digits = 2)),
marker = list(color = '#D95F02', size = 6, opacity = 0.65))
p3Create final trend plot by stacking with epidemic curve
smooth_extract <-
plotly::subplot(p3,p1, # plots to combine, top to bottom
nrows = 2,
heights = c(.6,.4), # relative heights of the two plots
shareX = TRUE, # plots will share an X axis
titleY = TRUE
) %>%
# create a vertical "spike line" to compare data across 2 plots
plotly::layout(
xaxis = list(
spikethickness = 1,
spikedash = "dot",
spikecolor = "black",
spikemode = "across+marker",
spikesnap = "cursor"
),
yaxis = list(spikethickness = 0)
)## Warning: Ignoring 1 observationssmooth_extractsave(smooth_extract, file = "./smooth_extract.rda")